4:51 PM
Mar 23, 2017
 |  BC Innovations  |  Tools & Techniques

The phenomics phenomenon

Why phenotypic screens are making a comeback

Having been upstaged by the dawning of genomics over 20 years ago, phenotypic screening is making a comeback as improvements in disease models, advances in microscopy and the rise of machine learning are making it easier to measure disease-relevant phenotypes and track down the targets responsible.

The dominant paradigm in drug discovery centers on identifying a target, often through genetic screening, and then working out how to drug it.

But many in industry believe that strategy hasn't translated well into the clinic and are opting to attack the problem from the other direction, by first screening compounds to find one that corrects a disease phenotype and then deconvoluting back to the target.

"In the past, taking the molecular approach we might have tried to pick a target of high risk and pursue it -- a very hard way to go. What we now readily consider is how to begin with the phenotype, turn it into a relevant assay, and then see what sort of readouts we can establish to enable small molecule screening," said Peter Tummino, VP of lead discovery at Johnson & Johnson’s Janssen Pharmaceutica N.V.

J&J is one of many companies phasing in phenotypic screening. Paul Andrews, director of operations at the U.K.'s National Phenotypic Screening Centre (NPSC), told BioCentury that pharma had "a change in mood" after several analyses published on the origins of new medicines surprisingly showed that most came from phenotypic studies rather than the target driven approaches dominating the industry.

"The hit rate in phenotypic screening is much much lower, at least 10-fold if not 100-fold lower. But what you get out usually is relevant and has an effect," said Andrews.

Traditional screening protocols involve step-by-step analyses after a target is identified, using one-dimensional biochemical assays such as receptor binding, or cell line readouts such as proliferation.

By contrast, phenotypic screens -- sometimes referred to as phenomics -- typically involve measuring a wide variety of cellular parameters in complex disease models that are more physiologically relevant (see "More is More").

Figure: More is more

Traditional drug screens are typically single parameter in vitro or cell line assays (gray box, left) that measure enzyme activity, protein binding or cell properties like proliferation or migration, but screens in drug development are evolving to measure a greater number of parameters (vertical axis) and to incorporate more complex screening models (horizontal axis).

Complex cellular models of human disease include co-cultures of two cell types, induced pluripotent stem (iPS) cell derived cell types, and multi-well or microfluidic organoid cultures. Readouts from the various 3-D cell models (Pink box, center) include protein and organelle marker expression measured using staining techniques like Cell Painting developed at the Broad Institute of MIT and Harvard and cell imaging analysis using software programs like Anne Carpenter's Cell Profiler. Current assays commonly measure more than 1,000 data points and can simultaneously analyze as many as eight different cell structures using protocols like Cell Painting. Carpenter is director of the imaging platform at Broad.

Animals such as flies, worms, and zebrafish can be used to generate genetic models of disease and provide mutant cells for phenotypic screens or potentially for organismal drug screens. Companies like Perlara PBC are generating cellular disease models from mutant flies, worms and fish and Teleos Therapeutics has screened compounds in zebrafish...

Read the full 2759 word article

User Sign in

Trial Subscription

Get a 4-week free trial subscription to BioCentury Innovations

Article Purchase

$100 USD
More Info >